Use this URL to cite or link to this record in EThOS:
Title: An analysis of the determinants of student dropout behaviour : the case of further education sector in the UK
Author: Capuccinello, Rossella Iraci
Awarding Body: Lancaster University
Current Institution: Lancaster University
Date of Award: 2012
Availability of Full Text:
Access from EThOS:
This thesis analyses the determinants of students’ dropout behaviour with a fo­cus on the further education sector in England. In particular, we give special atten­tion to the effect of institutional size and of college mergers on students’ probability of withdrawal, in an attempt to fill a gap in the literature on students’ outcomes. This work is composed of three empirical chapters. The first one focuses on the empirical analysis of the determinants of full and partial dropout behaviour and, in particular, on the effect of college and programme area size. We find that the re­lationship between partial and complete dropout and college and programme area size takes the shape of an inverted U. However, the gradient is quite small. This finding shows that medium sized colleges are somewhat less effective in retaining students. The second empirical chapter investigates the causal effect of enrolment in a recently merged further education institution on the probability of dropping out. We employ a propensity score matching approach. Therefore, we match the stu­dents enrolled on a merged college with students from the control group which are similar to them in terms of propensity score. Through this technique we find that enrolling on a recently merged institution reduces the probability of dropping out by about 1.6 to 4 percentage points. The final empirical chapter investigates whether the hazard of dropping out for both male and female students changes over the duration of study. Using duration modelling techniques degree of negative duration dependence for both males and females. However this pattern is sensitive to attempts to control for unobserved het­erogeneity. These extended models show a flattened hazard function especially for females suggesting that the hazard is basically constant over time. For males intro­ducing controls for unobserved heterogeneity also flattens the hazard, nonetheless there is still evidence of a heightened risk of dropping out in the first few months of study. In addition, we examine variations in drop out hazard patterns for students en­rolled on courses which confer different qualification levels. We provide evidence of distinct hazard patterns between students pursuing ’high level’ and ’low level’ qualifications.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available